INS (AE 555)

INS (555) GPS Segments

The total GPS configuration is comprised of three distinct segments:
ƒ The Space Segment. Satellites orbiting the earth.
ƒ The Control Segment. Stations positioned on the earth's equator to control the
GLOBAL POSITIONING SYSTEM (GPS) satellites
ƒ The User Segment. Anybody that receives and uses the GPS signal.
Introduction
The Global Positioning System is a worldwide radio-navigation system satellite-based
navigation system developed by the U.S. Department of Defense (DoD) formed from a
constellation of 24 satellites and their ground stations.
Fig.2 GPS segments
The Space Segment
The Space Segment is designed to consist of 24 satellites orbiting the earth at approximately
20200km every 12 hours. At time of writing there are 26 operational satellites orbiting the
Fig. 1 GPS Satellite Constellation earth. The space segment is so designed that there will be a minimum of 4 satellites visible
above a 15° cut-off angle at any point of the earth's surface at any one time. Four satellites
GPS uses these "man-made stars" as reference points to calculate positions accurate to a are the minimum that must be visible for most applications. Experience shows that there
matter of meters. GPS receivers have been miniaturized to just a few integrated circuits and are usually at least 5 satellites visible above 15° for most of the time and quite often there
so are becoming very economical. And that makes the technology accessible to virtually are 6 or 7 satellites visible. Each GPS satellite has several very accurate atomic clocks on
everyone. These days GPS is finding its way into cars, boats, planes, construction board. The clocks operate at a fundamental frequency of 10.23MHz. This is used to generate
equipment, movie making gear, farm machinery, and laptop computers. Soon GPS will the signals that are broadcast from the satellite. The satellites broadcast two carrier waves
become almost as basic as the telephone. GPS is the shortened form of NAVSTAR GPS. This constantly. These carrier waves are in the L-Band (used for radio), and travel to earth at the
is an acronym for Navigation System with Time and Ranging Global Positioning System. speed of light. These carrier waves are derived from the fundamental frequency, generated
After the Second World War, it became apparent to the U.S. Department of Defense that a by a very precise atomic clock:
solution had to be found to the problem of accurate, absolute positioning. Several projects
and experiments ran. All of these projects allowed positions to be determined but were
limited in accuracy or functionality. At the beginning of the 1970s, a new project was The Control Segment
proposed. This concept promised to fulfill all the requirements of the US government, The Control Segment consists of one master control station, 5 monitor stations and 4
namely that one should be able to determine ones position accurately, at any point on the ground antennas distributed amongst 5 locations roughly on the earth's equator. The
earth's surface, at any time, in any weather conditions. GPS was originally designed for Control Segment tracks the GPS satellites, updates their orbiting position and calibrates and
military use. Soon after the original proposals were made, it became clear that civilians synchronizes their clocks. A further important function is to determine the orbit of each
could also use GPS, and not only for personal positioning (as was intended for the military). satellite and predict its path for the following 24 hours. This information is uploaded to each
The first two major civilian applications to emerge were marine navigation and surveying. satellite and subsequently broadcast from it. This enables the GPS receiver to know where
Nowadays applications range from in car navigation through truck fleet management to each satellite can be expected to be found. The satellite signals are read at Ascension, Diego
automation of construction machinery. Garcia and Kwajalein.
 The problem with GPS is that only pseudo ranges and the time at which the signal arrived at
the receiver can be determined. Thus there are four unknowns to determine; position (X, Y,
Z) and time of travel of the signal. Observing to four satellites produces four equations
which can be solved, enabling these unknowns to be determined. At least four satellites are
required to obtain a position and time in 3 dimensions. In order to calculate the distance to
each satellite, one of Newton’s laws of motion is used:
Distance = Velocity x Time
For instance, it is possible to calculate the distance a train has traveled if you know the
velocity it has been traveling and the time for which it has been traveling at that velocity.
GPS requires the receiver to calculate the distance from the receiver to the satellite. The
Fig.3 Control Segment Station Locations Velocity is the velocity of the radio signal. Radio waves travel at the speed of light, 290,000
km per second (186,000 miles per second). The Time is the time taken for the radio signal to
travel from the satellite to the GPS receiver. This is a little harder to calculate, since you
The measurements are then sent to the Master Control Station in Colorado Springs where
need to know when the radio signal left the satellite and when it reached the receiver.
they are processed to determine any errors in each satellite. The information is then sent
back to the four monitor stations equipped with ground antennas and uploaded to the
satellites. Calculating the Time
The satellite signal has two codes modulated upon it, the C/A code and the P-code. The C/A
code is based upon the time given by a very accurate atomic clock. The receiver also
The User Segment contains a clock that is used to generate a matching C/A code. The GPS receiver is then able
The User Segment comprises of anyone using a GPS receiver to receive the GPS signal and
to match or correlate the incoming satellite code to the receiver generated code.
determine their position and/ or time. Typical applications within the user segment are land
The C/A code is a digital code that is .pseudo random or appears to be random. In actual
navigation for hikers, vehicle location, surveying, marine navigation, aerial navigation,
fact it is not random and repeats one thousand times every second. In this way, the time
machine control etc.
taken for the radio signal to travel from the satellite to the GPS receiver is calculated.
Satellite Ranging Error Sources
All GPS positions are based on measuring the distance from the satellites to the GPS There are several sources of error that degrade the GPS position from a theoretical few
receiver on the earth. This distance to each satellite can be determined by the GPS receiver. meters to tens of meters. These error sources are:
The basic idea is that of resection, which many surveyors use in their daily work. If you know
x Ionospheric and atmospheric delays
the distance to three points relative to your own position, you can determine your own
x Satellite and Receiver Clock Errors
position relative to those three points. From the distance to one satellite we know that the
x Multi path
position of the receiver must be at some point on the surface of an imaginary sphere which
x Dilution of Precision
has it's origin at the satellite. By intersecting three imaginary spheres the receiver position
x Selective Availability (S/A)
can be determined.
x Anti Spoofing (A-S)
™ Ionospheric and Atmospheric delays
As the satellite signal passes through the ionosphere, it can be slowed down, the
effect being similar to light refracted through a glass block. These atmospheric delays
can introduce an error in the range calculation as the velocity of the signal is
affected. (Light only has a constant velocity in a vacuum). The ionosphere does not
introduce a constant delay on the signal.
™ Satellite and Receiver clock errors
The clocks in the satellite are very accurate (about 3 n sec), but they do sometimes
drift slightly and cause small errors, affecting the accuracy of the position. The US
Fig.4. Intersection of three imaginary spheres Department of Defense monitors the satellite clocks using the Control Segment and
can correct any drift that is found.
™ Multi path Errors
Multi path occurs when the receiver antenna is positioned close to a large reflecting
Inertial Navigation System
surface such as a lake or building. The satellite signal does not travel directly to the
antenna but hits the nearby object first and is reflected into the antenna creating a Introduction
false measurement. Multi path can be reduced by use of special GPS antennas that
prevent low elevation signals reaching the antenna. Multi path only affects high Inertial Navigation Assembly (ISA) is an ensemble of inertial sensors rigidly mounted to a
accuracy, survey type measurements. Simple handheld navigation receivers do not common base to maintain the same relative orientations. Inertial sensor assemblies used in
employ such techniques. inertial navigation usually contain three accelerometers and three gyroscopes. Inertial
Measurement Unit (IMU) includes an ISA and its required electronics for the purpose of
calibration and control. The support electronics may also include thermal control or
compensation, signal conditioning, and input/output control. The IMU may also include an
IMU processor, and for gimbaled systems the gimbal control electronics INS is an IMU
however the output from the unit is fed to navigation algorithms so that the position,
velocity, attitude and heading of the vehicle can be evaluated. The unit also provides the
compensated raw data which can be used for control or stabilization purposes. INS systems
are generally found in almost all forms of long distance aircraft and sea vessels, submarine
and missile applications, and this is due to their initial wide spread use in military roles. In
such applications the inertial sensors implemented have to be of supreme quality, providing
stable readings, extremely high resolution and high-bandwidth. The algorithms and
electronics implemented are also of high quality in order to minimize the introduction of
any errors. With the current trend to better navigation performance for civilian applications,
Fig.5 Multi path effect. INS systems can provide a useful sensor; however, current high accuracy INS systems are
™ Dilution of Precision too expensive for land vehicle or robotic applications.
The Dilution of Precision (DOP) is a measure of the strength of satellite geometry and
is related to the spacing and position of the satellites in the sky. The DOP can
magnify the effect of satellite ranging errors. The range to the satellite is affected by
range errors previously described Different types of Dilution of Precision or DOP can
be calculated depending on the dimension.
™ Selective Availability (S/A)
Selective Availability is a process applied by the U.S. Department of Defense to the
GPS signal. This is intended to deny civilian and hostile foreign powers the full
accuracy of GPS by subjecting the satellite clocks to a process known as dithering
which alters their time slightly. Additionally, the ephemeris (or path that the satellite
will follow) is broadcast as being slightly different from what it is in reality. The end
result is degradation in position accuracy. It is worth noting that S/A affects civilian The basic principle of an INS is based on the integration of accelerations observed by the
users using a single GPS receiver to obtain an autonomous position. Users of accelerometers on board the moving platform. The system accomplishes this task through
differential systems are not significantly affected by S/A. Currently, it is planned that appropriate processing of the data obtained from the specific force and angular velocity
S/A will be switched off by 2006 at the latest. measurements. Thus, an appropriately initialized inertial navigation system is capable of
continuous determination of vehicle position, velocity and attitude without the use of the
™ Anti-Spoofing (A-S) external information.
Anti-Spoofing is similar to S/A in that it's intention is to deny civilian and hostile A major advantage of using inertial units is that given the acceleration and angular rotation
powers access to the P-code part of the GPS signal and hence force use of the C/A rate data in three dimensions, the velocity and position of the vehicle can be evaluated in
code which has S/A applied to it. Anti-Spoofing encrypts the P-code into a signal any navigation frame. However, the errors caused by bias, scale factors and non-linearity in
called the Y-code. Only users with military GPS receivers (the US and its allies) can the sensor readings cause an accumulation in navigation errors with time and furthermore
de-crypt the Y-code. inaccurate readings are caused by the misalignment of the unit's axes with respect to the
local navigation frame. Since an inertial unit is a dead reckoning sensor, any error in a
previous evaluation will be carried onto the next evaluation, thus as time progresses the Coincides with sensor frame. The Z-axis is orthogonal to the reference ellipsoid pointing up.
navigation solution drifts. The X-axis is pointing towards geodetic East. The Y-axis is pointing toward geodetic North.
The first type of INS developed was a gimbaled system. The accelerometers are mounted on The body frame represents the orientation of the IMU axes. The IMU sensitive axes are
a motorized gimbaled platform which was always kept aligned with the navigation frame. assumed to be approximately coincident with the moving platform upon which the IMU
Pickups are located on the outer and inner gimbals which keep track of the attitude of the sensors are mounted. The origin of the body frame is at the center of the IMU. The X-axis
stabilized platform relative to the vehicle on which the INS is mounted. These systems are points towards the right of the moving platform, the Y-axis points toward the front of the
very accurate, because the sensors can be designed for very precise measurements in a moving platform, and the Z-axis is orthogonal to the X and Y axes to complete a right-
small measurement range. However this setup has several detractors which make it handed frame.
undesirable. In contrary, a strap-down inertial navigation system uses orthogonal
accelerometers and gyro triads rigidly fixed to the axes of the moving vehicle. The angular
motion of the system is continuously measured using the rate sensors. The accelerometers Physical implementation
do not remain stable in space, but follow the motion of the vehicle. A strap-down system is
a major hardware simplification of the old gimbaled systems. The accelerometers and gyros The first type of INS developed was a gimbaled system the accelerometers are mounted on
are mounted in body coordinates and are not mechanically moved. Instead, a software a gimbaled. In a gimbaled system the accelerometer triad is rigidly mounted on the inner
solution is used to keep track of the orientation of the IMU (and vehicle) and rotate the gimbal of three gyros. The inner gimbal is isolated from the vehicle rotations and its attitude
measurements from the body frame to the navigational frame. This method overcomes the remains constant in a desired orientation in space during the motion of the system. The
problems encountered with the gimbaled system, and most importantly reduces the size, gyroscopes on the stable platform are used to sense any rotation of the platform, and their
cost, power consumption, and complexity of the system. outputs are used in servo feedback loops with gimbal pivot torque actuators to control the
gimbals such that the platform remains stable. These systems are very accurate, because
the sensors can be designed for very precise measurements in a small measurement range.
Coordinate Frame Definition
Three coordinate frames are important for this work. These include the ECEF (Earth-
Centered Earth-Fixed) frame (e frame), the body frame (b frame) and the local level frame
(LLF). The three frames are shown in Figure 3. The origin of the ECEF frame is the center of
the Earth’s mass. The X-axis is located in the equatorial plane and points towards the mean
Meridian of Greenwich. The Y– axis is also located in the equatorial plane and is 90 degrees
east of the mean Meridian of Greenwich. The Z-axis parallels the Earth’s mean spin axis. LLF
is a local geodetic frame serves as local reference directions for representing vehicle
attitude and velocity for operation on or near the surface of the Earth; for this reason, it is
often referred to as navigation frame (n-frame). A common orientation for LLF coordinates
is the North-East-Up (NEU) system. The origin of the LLF frame is
In contrary, a strap-down inertial navigation system uses orthogonal accelerometers and
gyro triads rigidly fixed to the axes of the moving vehicle (figure 2.a). The angular motion of
the system is continuously measured using the rate sensors. The accelerometers do not
remain stable in space, but follow the motion of the vehicle. A strap-down system is a major
hardware simplification of the old gimbaled systems. The accelerometers and gyros are
mounted in body coordinates and are not mechanically moved. Instead, a software solution
is used to keep track of the orientation of the IMU (and vehicle) and rotate the
measurements from the body frame to the navigational frame. This method overcomes the
problems encountered with the gimbaled system, and most importantly reduces the size,
cost, power consumption, and complexity of the system.
Error sources
This section will provide a quick overview of some difficulties present in inertial navigation. Two strapdown inertial navigation
This will provide a better understanding for the difficulties encountered with the IMU.
For a vehicle moving in 2D space, it is necessary to monitor both the translational motion in
x Bias. A sensor bias is always defined by two components: A deterministic component two directions and the change in the direction of vehicle (i.e. rotational motion). Two
called bias offset which refers to the offset of the measurement provided by the accelerometers are required to detect the acceleration in two directions. One gyroscope is
sensor from the true input; and a stochastic component called bias drift which refers required to detect the direction of the vehicle (rotational motion) in a direction
to the rate at which the error in an inertial sensor accumulates with time. The bias perpendicular to the plane of motion. Strapdown systems mathematically transform the
offset is deterministic and can be quantified by calibration while the bias drift is output of the accelerometers attached to the body into the navigation coordinate system
random in nature and should be treated as a stochastic process. before performing the mathematical integration. These systems use the output of the
gyroscope attached to the body to continuously update the transformation necessary to
x The scale factor. Is the relationship between the output signal and the true physical convert from body coordinate to navigation. The derivation of the transformation matrix is
quantity being measured and it is usually expressed in parts per million (ppm). The explained below.
scale factor is deterministic in nature and can be quantified or determined through
lab calibration. The variation of the scale factor with the variation of the exerted
acceleration/angular rate or temperature represents the scale factor stability and is
usually called the non-linear part of the scale factor error.
x Output stability. The output stability of a sensor defines the run-to-run or switch-on-
to-switch-on variation of the gyro-drift/accelerometer-bias as well as in-run variation
of gyro-drift/accelerometer-bias. The run-to-run stability can be evaluated from the
scatter in the mean output for each run for a number of runs given that the sensor is
turned off then on again between each two successive runs. The in-run stability of a
sensor is deduced from the average scatter of the measured drift in the output
about the mean value during a single run.
AS seen from the above figure the two accelerometers are fixed in X and Y directions, these
x Thermal sensitivity. Thermal sensitivity refers to the range of variation of the sensor directions represent the body coordinates. The measured acceleration will transform to the
performance characteristics, particularly bias and scale factor errors, with a change navigation frame (East and North) using the following transformation matrix.
in temperature. A bias or scale factor correlation with temperature variation can be
defined graphically or numerically (using a mathematical expression) through (5)
intensive lab thermal testing. Such correlations can be stored on a computer for
online use to provide compensation for temperature variation, provided a thermal
(6)
sensor is supplied with the sensor.
Where
Effect of Inertial Sensor Errors in Navigation Parameters
: are the accelerations in the (East and North directions) navigation frame.
A: is azimuth angle.
: is the acceleration in the navigation frame.
: is the acceleration in the body frame defined by the accelerometers.
: is the rotation matrix which rotates to the navigation frame.
Inertial navigation mechanization equations in 2D.
INS mechanization is the process of determining the navigation states (position, velocity and
attitude) from the raw inertial measurements through solving the differential equations
An accelerometer bias introduces a linear error in velocity and a quadratic error in the
describing the system motion. Mechanization is usually expressed by a set of differential
position.
equations and typically performed in the local level frame defined by the local east, north There for, equation (7) can be rewritten as follow
and ellipsoid normal. The IMU measurements include one angular rate components (14)
provided by the gyroscope and denoted by as well as two linear acceleration components
provided by the accelerometers and denoted by the 2x1 vector . This means that the (15)
angular velocity of the body frame is measured with respect to the inertial frame.
The term in equation (15) represents the angular acceleration induced due to the
relative rotation of the body frame with respect to the navigation frame. Finally the
vb acceleration components in the navigation frame can be expressed as
Rbn  vl
X_position
- (16)
Acceleration vl
Integration
CAR IMU 
Integration
GPS/INS Integration
Steering
Y_position
ȍ GPS and INS have complementary qualities that make them ideal use for sensor fusion. The
Integration
limitations of GPS include occasional high noise content, outages when satellite signals are
ș
blocked, interference and low bandwidth. The strengths of GPS include its long-term
stability and its capacity to function as a stand-alone navigation system. In contrast, inertial
navigation systems are not subject to interference or outages, have high bandwidth and
Considering the block diagram shown in figure 5, the differential equations describing the good short-term noise characteristics, but have long-term drift errors and require external
INS mechanization equation can be derived as follows; Firstly, the output of the two information for initialization. A combined system of GPS and INS subsystems can exhibit the
accelerometers is transformed from the body frame to the navigation frame (local level robustness, higher bandwidth and better noise characteristics of the inertial system with the
frame) using the transformation matrix as given in the following equation long-term stability of GPS.
(7) Comparison between INS and GPS
Where
Advantages Disadvantages
(8) 1- High position, velocity, and attitude 1- Bias drift problems (accuracy deteriorates
accuracy over short term with time)
2- High measurement output rate 2- Requires knowledge of gravity
After the transformation of the acceleration components from the body frame to the (much bigger than GPS) 3- Requires initial and in-flight calibration and
navigation frame, the velocity components can be derived by integrating the given 3- Autonomous system, no external alignment
acceleration in the navigation frame. Consequently the corresponding position can be electromagnetic signals (non-
INS
obtained by double integrating the acceleration in the navigation frame. Also the velocity jamming)
component of the navigation frame can be directly calculated using the transformation of 4- Works in all environments (indoor,
the velocity component in the body frame using the transformation matrix underwater, and underground)
5- Excellent performance in high
dynamic environment
(9)
6- Exhibits relatively low noise
1- Accurate specially for DGPS using 1- Susceptibility to jamming
Differentiating equation (9) with respect to time, yields carrier measurements 2- Requirement for line-of-sight between the
2- Doesn’t have any bias drift error receiver antenna and the satellites, which
(10) (long-term accuracy) cannot always be met
Where 3- Provides deterministic solution for 3- Poor perform in high dynamic environment
(11) GPS position and velocity when there is high rate of change of
ɏŝƐĂƐŬĞǁƐLJŵŵĞƚƌŝĐŵĂƚƌŝdžĚĞĨŝŶĞĚĂƐ 4- Not sensitive to gravity Doppler
(12) 4- Data rate is not sufficient for precise
trajectory determination
Then 5- Not have of attitude information
(13) 6- Relatively noisy with respect to INS